Method for calculating infrared radiation intensity of point source target in air

Abstract

The invention discloses a method for calculating the infrared radiation intensity of an aerial point source target, and the method comprises the steps: calibrating an infrared detector, determining the response gain coefficient and the pixel offset of each pixel of an infrared image through the calibration, and constructing a response gain coefficient matrix and a pixel offset matrix; determining the radiation brightness of each pixel point of the aerial measured point source target projection area image according to the response gain coefficient and the pixel offset; the step of determining the projection area of the detected point source target in the air is an effective projection area of the detected point source target in the air, the point source target infrared radiation intensity calculation method is improved, the precision of the calculation method is improved, and the calculation efficiency is improved. The method has the characteristics of high operability and wide applicability, and can be used as a universal method for calculating the infrared radiation intensity of the aerial point source target.

Description

Technical Field

[0001] The invention relates to a method for calculating the infrared radiation intensity of an aerial point source target. Background Art

[0002] In the long-distance detection of targets and the estimation of the limit range, we often encounter small infrared targets. At this time, the target is far away from the infrared measurement system. In theory, the solid angle of the target to the infrared system is smaller than the solid angle of a single pixel of the detector. However, due to the diffraction of the optical system and the random aberration of the atmosphere, the target appears as a small light spot with diffuse energy on the target surface of the detector. The target in this case is also called a point source target. The infrared characteristics of point source targets are generally described by radiation intensity. Radiation intensity refers to the radiation flux emitted by a point radiation source within a unit solid angle in a specific direction. It is usually used to describe the strength of the infrared radiation energy of a point source target in a specific direction.

[0003] The traditional process of calculating the radiation intensity of a point source target is generally divided into three steps. First, the grayscale value of the infrared system output image is inverted into the radiation brightness value through the infrared radiation calibration coefficient calculated by calibration. Then, the projection area of ​​the point source target on the detector target surface is calculated to obtain the number of effective imaging pixels of the target. Finally, the radiation intensity of the point source target is obtained by integrating the corresponding radiation brightness of each effective imaging pixel. There are two main factors that affect the final calculation result: one is the infrared system calibration coefficient, including the response gain coefficient and bias. On the one hand, the traditional method assumes that the response degree of each pixel of the sensor is equal, ignoring the individual slight differences caused by factors such as production process. On the other hand, the experiment found that the sensor bias coefficient will change with temperature, and the overall linear correlation is formed. In actual measurement, most of them do not take real-time correction. The second is the number of effective imaging pixels of the target. Because the energy of the point source target projected to the sensor is weak, it is easy to be covered by the background radiation of the sky, resulting in the inability to accurately select the true imaging size of the target. The grayscale threshold selection method used in the traditional method also does not consider the system error caused by the sensor. Summary of the invention

[0004] The purpose of the present invention is to propose a method for calculating the infrared radiation intensity of an aerial point source target. First, the infrared radiation is calibrated for each pixel on the detector target surface, and a response gain coefficient matrix and a bias matrix are established. The grayscale image generated by the detector is converted into a radiation brightness image. Then, effective imaging pixel judgment of the point target based on the 3σ principle of radiation brightness is carried out, and the radiation brightness of the target area is obtained by background cancellation, and the influence of temperature on the bias is eliminated. Finally, the target area is integrated to obtain the target infrared radiation intensity.

[0005] In order to achieve the above object, the technical solution of the present invention is:

[0006] A method for calculating the infrared radiation intensity of an aerial point source target is used for measuring infrared radiation characteristics. The method comprises an infrared detector. The calculation comprises: calibrating the infrared detector, determining the response gain coefficient and pixel offset of each pixel of the infrared image by calibration, and constructing a response gain coefficient matrix and a pixel offset matrix; determining the radiation brightness of each pixel point of the projection area image of the aerial point source target to be measured by the response gain coefficient and the pixel offset; integrating the radiation brightness of each pixel point of the target projection area to obtain the infrared radiation intensity of the target area, wherein: determining the projection area of ​​the aerial point source target to be measured is to determine the effective projection area of ​​the aerial point source target to be measured, and the process is:

[0007] The measured point source target is stably tracked to obtain an infrared grayscale image including the measured target and the pure sky background. The target grayscale image is converted pixel by pixel into a radiation brightness image using a response gain coefficient matrix and a pixel offset matrix. The target area in the radiation brightness image is divided into area A, area B and area C. Area A includes all point source targets, with a total of N1 pixels. Area B is an area where area A extends outward by multiple pixels, with a total of N2 pixels. The matrix ring area C inside area B and outside area A is the sky background, with a total of N2-N1 pixels. The pixels in area A of the infrared radiation brightness image whose radiation brightness value is greater than the background radiation brightness mean by at least 3 times the sum of the background radiation brightness standard deviations are defined as target effective imaging pixels. The area where the target effective imaging pixels exist is the target effective projection area D. The infrared radiation intensity of the target area is obtained by integrating the radiation brightness of each pixel point in the target effective projection area by the constructed target area infrared radiation intensity model.

[0008] The solution is further that: the infrared radiation intensity model of the target area is:

[0009]

[0010] in,

[0011]

[0012] H target (x, y) is the grayscale value of the target area D at (x, y);

[0013] is the average gray value of the sky background area C image;

[0014] is the average bias of the sky background area C;

[0015] B(x,y) Bias coefficients corresponding to pixels at different positions

[0016] τ atm is the measured atmospheric transmittance;

[0017] A d is the detector pixel size;

[0018] M s is the magnification of the infrared system;

[0019] G(x, y) is the response gain coefficient and pixel bias of the pixel at (x, y) in the sky background area C;

[0020] Δ is the apparent radiation intensity of the atmosphere outside the target slant path:

[0021]

[0022] N0 is the target effective pixel number;

[0023] L path is the atmospheric radiation;

[0024] L back is the measured sky background radiation;

[0025] is the calibrated average response gain coefficient of the sky background area C.

[0026] A further solution is: the region B is a region where the region A extends outward by multiple pixels, and is a region where the region extends outward by at least 3 pixels.

[0027] The scheme is further as follows: the response gain coefficient and pixel offset of each pixel of the infrared image are determined by calibration, and the response gain coefficient matrix and pixel offset matrix are constructed as follows:

[0028] The first step is to obtain the standard surface source black body image and measure the standard surface source black body at n different temperature points T i Carry out measurement and record the corresponding images H at different temperature points i , n≥2, i is an integer and 1≤i≤n;

[0029] The second step is to use the linear regression method to calculate the linear relationship between the grayscale value of each pixel response of the infrared image and the target infrared radiation brightness at different temperatures, and establish the response gain coefficient matrix G and the pixel bias matrix B;

[0030]

[0031]

[0032] in:

[0033] x and y represent the image H i The pixel position coordinates along the x-direction and the y-direction;

[0034] L i is the temperature Ti Theoretical radiation brightness of standard surface black body under ;

[0035] is L at n different temperature points i Mean;

[0036] H i (x, y) is the temperature T i The gray value of the pixel at (x, y) in the infrared mean image under;

[0037] G(x, y) and B(x, y) are the response gain coefficient and pixel bias of the pixel at (x, y), respectively.

[0038] The solution is further that: the radiation brightness of each pixel point in the target effective projection area is obtained by constructing a brightness model, and the brightness model is:

[0039]

[0040] in:

[0041] L target (x, y) is the radiation brightness value of the target area at (x, y), (x, y)∈D;

[0042] H target (x, y) is the grayscale value of the target area D at (x, y);

[0043] is the average gray value of the sky background area C image;

[0044] τ atm is the measured atmospheric transmittance;

[0045] L path is the atmospheric radiation;

[0046] L back is the measured sky background radiation;

[0047]

[0048] is the calibrated average response gain coefficient of the sky background area C;

[0049] is the average bias of the sky background area C;

[0050] B(x,y) is the bias coefficient corresponding to pixels at different positions;

[0051] G(x, y) and B(x, y) are the response gain coefficient and pixel bias of the pixel at (x, y) in the sky background area C, respectively.

[0052] The main advantages of the scheme of the present invention are: a method for calculating the infrared radiation intensity of an aerial point source target is proposed, and the linear relationship between the grayscale response of a single pixel and a standard infrared radiation source is obtained by calibrating the infrared radiation of each pixel of the detector, thereby improving the whole-frame grayscale mean calibration method commonly used in the current infrared characteristic test, solving the problem that the traditional calculation method has a large error in the result due to the non-uniformity of the response degree between different pixels of the detector, and judging the effective imaging pixels of the point target based on the principle of 3 times the radiation brightness, and calculating the target projection area, and finally eliminating the influence of temperature drift on the bias through background cancellation, thereby improving the infrared radiation intensity calculation method of the point source target, and improving the accuracy of the calculation method, having the characteristics of strong operability and wide applicability, and can be used as a universal infrared radiation intensity calculation method for aerial point source targets.

[0053] The present invention is described in detail below with reference to the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a schematic diagram of the process steps of the present invention;

[0055] Figure 2 Schematic diagram of the area where the infrared image target is located. DETAILED DESCRIPTION

[0056] A method for calculating the infrared radiation intensity of an aerial point source target is used for measuring infrared radiation characteristics. The method comprises an infrared detector. The calculation comprises: calibrating the infrared detector, determining the response gain coefficient and pixel offset of each pixel of the infrared image by calibration, and constructing a response gain coefficient matrix and a pixel offset matrix; determining the radiation brightness of each pixel point of the projection area image of the aerial point source target to be measured by the response gain coefficient and the pixel offset; integrating the radiation brightness of each pixel point of the target projection area to obtain the infrared radiation intensity of the target area, wherein: determining the projection area of ​​the aerial point source target to be measured is to determine the effective projection area of ​​the aerial point source target to be measured, and the process is:

[0057] The measured point source target is stably tracked to obtain an infrared grayscale image that includes both the measured target and the pure sky background. The target grayscale image is converted pixel by pixel into a radiance image using the response gain coefficient matrix and the pixel offset matrix. Figure 2As shown, the target area in the radiation brightness image is divided into area A, area B and area C. Area A contains all point source targets, with a total of N1 pixels; area B is an area that extends multiple pixels outward from area A, with a total of N2 pixels; the matrix ring area C inside area B and outside area A is the sky background, with a total of N2-N1 pixels, and the pixels in the infrared radiation brightness image area A whose radiation brightness value is greater than the background radiation brightness mean by at least 3 times the sum of the background radiation brightness standard deviations are defined as target effective imaging pixels. The area where the target effective imaging pixels exist is the target effective projection area D, and the infrared radiation intensity of the target area is obtained by integrating the radiation brightness of each pixel point in the target effective projection area by the constructed target area infrared radiation intensity model.

[0058] The infrared radiation intensity model of the target area is:

[0059]

[0060] in,

[0061]

[0062] H target (x, y) is the grayscale value of the target area D at (x, y);

[0063] is the average gray value of the sky background area C image;

[0064] is the average bias of the sky background area C;

[0065] B(x,y) is the bias coefficient corresponding to pixels at different positions;

[0066] τ atm is the measured atmospheric transmittance;

[0067] A d is the detector pixel size;

[0068] M s is the magnification of the infrared system;

[0069] G(x, y) is the response gain coefficient and pixel bias of the pixel at (x, y) in the sky background area C;

[0070] Δ is the apparent radiation intensity of the atmosphere outside the target slant path:

[0071]

[0072] N0 is the target effective pixel number;

[0073] L path is the atmospheric radiation;

[0074] L back is the measured sky background radiation;

[0075] is the calibrated average response gain coefficient of the sky background area C.

[0076] The region B is a region extending outwards from the region A by multiple pixels, and is a region extending outwards by at least 3 pixels.

[0077] The response gain coefficient and pixel offset of each pixel of the infrared image are determined by calibration, and the response gain coefficient matrix and pixel offset matrix are constructed as follows:

[0078] The first step is to obtain the standard surface source black body image and measure the standard surface source black body at n different temperature points T i Carry out measurement and record the corresponding images H at different temperature points i , n≥2, i is an integer and 1≤i≤n;

[0079] The second step is to use the linear regression method to calculate the linear relationship between the grayscale value of each pixel response of the infrared image and the target infrared radiation brightness at different temperatures, and establish the response gain coefficient matrix G and the pixel bias matrix B;

[0080]

[0081]

[0082] in:

[0083] x and y represent the image H i The pixel position coordinates along the x-direction and the y-direction;

[0084] L i is the temperature T i Theoretical radiation brightness of standard surface black body under ;

[0085] is L at n different temperature points i Mean;

[0086] H i (x, y) is the temperature T i The gray value of the pixel at (x, y) in the infrared mean image under;

[0087] G(x, y) and B(x, y) are the response gain coefficient and pixel bias of the pixel at (x, y), respectively.

[0088] The radiation brightness of each pixel point in the target effective projection area is obtained by constructing a brightness model, and the brightness model is:

[0089]

[0090] in:

[0091] L target (x, y) is the radiation brightness value of the target area at (x, y), (x, y)∈D;

[0092] H target (x, y) is the grayscale value of the target area D at (x, y);

[0093] is the average gray value of the sky background area C image;

[0094] τ atm is the measured atmospheric transmittance;

[0095] L path is the atmospheric radiation;

[0096] L back is the measured sky background radiation;

[0097]

[0098] is the calibrated average response gain coefficient of the sky background area C;

[0099] is the average bias of the sky background area C;

[0100] B(x,y) is the bias coefficient corresponding to pixels at different positions;

[0101] G(x, y) and B(x, y) are the response gain coefficient and pixel bias of the pixel at (x, y) in the sky background area C, respectively.

[0102] The above method is based on Figure 1 The described process steps perform the following operations:

[0103] First, step one: calibrate the infrared detector:

[0104] During the infrared calibration process, the standard surface source black body is calibrated at n different temperature points T i Carry out measurement and record the corresponding images H at different temperature points i (n≥2, i is an integer and 1≤i≤n);

[0105] The linear regression method is used to calculate the linear relationship between the grayscale value of each pixel response of the infrared image and the target infrared radiation brightness at different temperatures, and the response gain coefficient matrix G and the bias matrix B are established;

[0106]

[0107]

[0108] Among them, L i is the temperature T i The standard surface black body theoretical radiation brightness under is L at n different temperature points i Mean, H i (x, y) is the temperature T i The gray value of the pixel at (x, y) in the infrared mean image under the condition, G(x, y) and B(x, y) are the response gain coefficient and bias of the pixel at (x, y) respectively;

[0109] Step 2: Stably track the measured point source target to obtain an infrared grayscale image that includes both the measured target and the pure sky background. Use the detection response gain coefficient matrix G and the bias matrix B to convert the target grayscale image pixel by pixel into a radiation brightness image L0;

[0110] The target area in the infrared radiation brightness image of the detector is divided into area A, area B and area C. Area A contains all point source targets and part of the sky background, with a total of N1 pixels. Area B is the outer expansion area of ​​area A, generally extending outward by 3 pixels, with a total of N2 pixels. Among them, the matrix ring area C inside area B and outside area A is the sky background, with a total of N2-N1 pixels, and its background radiation brightness average is The standard deviation of the background radiation brightness σ b It is expressed as:

[0111]

[0112]

[0113] The infrared radiation brightness image area A has a radiation brightness value L0(x, y) greater than the background radiation brightness average 3 times the standard deviation of background radiation brightness σ b The pixel of the sum is defined as the target effective imaging pixel. F(x, y) is defined to indicate whether the pixel at (x, y) is a valid pixel. Valid is 1 and invalid is 0.

[0114]

[0115] The target projection area is the area where the effective imaging pixels of the target exist, which is defined as area D. The number of effective pixels of the target is represented by N0:

[0116] N0=∑F(x,y)(x,y)∈A (6)

[0117] Step 3: Define H target(x, y) is the grayscale value of the target area D at (x, y), and is defined as is the average gray value of the sky background area C image, and are the calibrated average response gain coefficient and average bias of the sky background area C, respectively, and H(x, y) is the grayscale value of the target infrared image recorded by the detector at (x, y):

[0118]

[0119]

[0120] The radiation brightness value L of the target at (x, y) target (x, y) can be expressed as, where (x, y)∈D:

[0121]

[0122] in, L target (x, y) is the radiation brightness value of the target at (x, y), τ atm is the measured atmospheric transmittance, L path is the atmospheric radiation, L back is the measured sky background radiation.

[0124] Integrate the pixels in the target area D to obtain the infrared radiation intensity I of the point source target target , where (x,y))∈D:

[0125]

[0126] Among them, A d is the detector pixel size, M s is the infrared system magnification, Δ is the apparent radiation intensity of the atmosphere outside the target slant path:

[0127]

[0128] In the calibration calculation formula of the embodiment, the bias matrix B(x, y) is regarded as a quantity related to the performance of the infrared system pixel itself. The bias coefficients corresponding to pixels in different positions are different. However, in the actual calibration process, it is found that the system bias coefficient is affected by the ambient temperature and changes linearly with the change of the ambient temperature. Moreover, the change of the bias coefficient B(x, y) corresponding to different pixels is basically the same.

[0129] Therefore, the linear relationship between bias B and ambient temperature is set to B = aT t +b, a is the linear coefficient, b is the linear bias, then the relationship between the bias coefficient B(x, y) corresponding to the pixels at different positions and the ambient temperature is B(x, y)) = aTt +b(x,y)

[0130] This means that in a linear relationship, the linear coefficient a is independent of (x, y), and the linear bias b is related to (x, y). is the average bias of the sky background area C. The average here is the average of all (x, y), so it is reflected in the linear relationship with the ambient temperature as Among them, a has nothing to do with (x, y) so it does not change, and b is related to (x, y) so we take the average value

[0131] The above-mentioned calculation method embodiment obtains the linear relationship between the grayscale response of a single pixel and the standard infrared radiation source by calibrating the infrared radiation of each pixel of the detector, thereby improving the whole-frame grayscale mean calibration method commonly used in the current infrared characteristic test, and solves the problem of large result errors caused by the non-uniformity of the response degree between different pixels of the detector in the traditional calculation method. The effective imaging pixels of the point target are judged based on the principle of 3 times the radiation brightness, and the target projection area is calculated. Finally, the influence of temperature drift on the bias is eliminated through background cancellation, and the calculation method of the infrared radiation intensity of the point source target is improved, and the accuracy of the calculation method is improved. It has the characteristics of strong operability and wide applicability, and can be used as a general infrared radiation intensity calculation method for aerial point source targets.

Claims

1. A method for calculating the infrared radiation intensity of an aerial point source target, used for measuring infrared radiation characteristics, includes an infrared detector, and the calculation includes: Calibrate the infrared detector, determine the response gain coefficient and pixel offset of each pixel of the infrared image through calibration, and construct the response gain coefficient matrix and pixel offset matrix; The radiation brightness of each pixel of the image of the projection area of ​​the measured point source target in the air is determined by the response gain coefficient and the pixel offset; the radiation brightness of each pixel of the target projection area is integrated to obtain the infrared radiation intensity of the target area, characterized in that the determination of the projection area of ​​the measured point source target in the air is to determine the effective projection area of ​​the measured point source target in the air, and the process is: The measured point source target is stably tracked to obtain an infrared grayscale image including the measured target and the pure sky background. The target grayscale image is converted pixel by pixel into a radiation brightness image using a response gain coefficient matrix and a pixel offset matrix. The target area in the radiation brightness image is divided into area A, area B and area C. Area A includes all point source targets, with a total of N1 pixels. Area B is an area where area A extends outward by multiple pixels, with a total of N2 pixels. The matrix ring area C inside area B and outside area A is the sky background, with a total of N2-N1 pixels. The pixels in area A of the infrared radiation brightness image whose radiation brightness value is greater than the background radiation brightness mean by at least 3 times the sum of the background radiation brightness standard deviations are defined as target effective imaging pixels. The area where the target effective imaging pixels exist is the target effective projection area D. The infrared radiation intensity of the target area is obtained by integrating the radiation brightness of each pixel point in the target effective projection area by the constructed target area infrared radiation intensity model.

2. The calculation method according to claim 1, characterized in that: The infrared radiation intensity model of the target area is: in, H target (x, y) is the grayscale value of the target area D at (x, y); is the average gray value of the sky background area C image; is the average bias of the sky background area C; B(x, y) is the bias coefficient corresponding to pixels at different positions; τ atm is the measured atmospheric transmittance; A d is the detector pixel size; M s is the magnification of the infrared system; G(x, y) is the response gain coefficient and pixel bias of the pixel at (x, y) in the sky background area C; Δ is the apparent radiation intensity of the atmosphere outside the target slant path: N0 is the target effective pixel number; L path is the atmospheric radiation; L back is the measured sky background radiation; is the calibrated average response gain coefficient of the sky background area C.

3. The calculation method according to claim 1, characterized in that: The region B is a region extending outwards from the region A by multiple pixels, and is a region extending outwards by at least 3 pixels.

4. The calculation method according to claim 1, characterized in that: The response gain coefficient and pixel offset of each pixel of the infrared image are determined by calibration, and the response gain coefficient matrix and pixel offset matrix are constructed as follows: The first step is to obtain the standard surface source black body image and measure the standard surface source black body at n different temperature points T i Carry out measurement and record the corresponding images H at different temperature points i , n≥2, i is an integer and 1≤i≤n; The second step is to use the linear regression method to calculate the linear relationship between the grayscale value of each pixel response of the infrared image and the target infrared radiation brightness at different temperatures, and establish the response gain coefficient matrix G and the pixel bias matrix B; in: X and y represent the image H i The pixel position coordinates along the x and y directions; L i is the temperature T i Theoretical radiation brightness of standard surface black body under ; is L at n different temperature points i Mean; H i (x, y) is the temperature T i The gray value of the pixel at (x, y) in the infrared mean image under; G(x, y) and B(x, y) are the response gain coefficient and pixel bias of the pixel at (x, y), respectively.

5. The calculation method according to claim 1, characterized in that: The radiation brightness of each pixel point in the target effective projection area is obtained by constructing a brightness model, and the brightness model is: in: L target (x, y) is the radiation brightness value of the target area at (x, y), (x, y)∈D; H target (x, y) is the grayscale value of the target area D at (x, y); is the average gray value of the sky background area C image; τ atm is the measured atmospheric transmittance; L path is the atmospheric radiation; L back is the measured sky background radiation; is the calibrated average response gain coefficient of the sky background area C; is the average bias of the sky background area C; B(x, y) is the bias coefficient corresponding to pixels at different positions; G(x, y) and B(x, y) are the response gain coefficient and pixel bias of the pixel at (x, y) in the sky background area C, respectively.

Images